A3010: Bodge wires

Arc/RPCs, peripherals, RISCOS operating system & ARM kit eg GP2x, BeagleBoard
philpem
Posts: 6
Joined: Fri Apr 04, 2014 5:42 pm

A3010: Bodge wires

Postby philpem » Mon Sep 11, 2017 12:03 am

Hi,

On the A3010 "0194,004 Issue 1" mainboard, there are two bodge wires:
* one from somewhere under the ID chip to a diode and capacitor connected to a pad below the ARM250TG's clock crystal
* another from a via near L14 through a resistor to somewhere near the battery or PL9 (floppy connector)

Sadly on mine, the second bodge wire (with the resistor) has come detached from PL9.
Does anyone know where this is supposed to be connected?

(for bonus points, does anyone know what issues these bodge-wires fix?)

This machine has suffered a battery leak and the damned battery goop has eaten through this particular bodge wire!

(If anyone remembers my thread some time ago about the '3010 which had a battery leak and started throwing RAM Control Line failures... this is the same machine. I'm having another go at fixing it before I scrap it off)

Cheers,
Phil.

BradBrad
Posts: 39
Joined: Sat Sep 02, 2017 9:40 pm

Re: A3010: Bodge wires

Postby BradBrad » Tue Sep 26, 2017 1:18 am

Not sure. Just did an A3000 refurb which had pretty bad battery damage but I think it is a totally different layout to an A3010? Maybe you could get a clue where it came off by seeing if the wire left any solder/indent on what it used to be connected to, combined with following the lay of the wire? Seems a shame to bin it for the sake of 1 wire.

If its just something to do with the battery or floppy, can it run without the wire? As long as it is nothing to do with the cpu clock. I seem to recall reading that running it without a clock for more than a few seconds can screw the ARM chip? :shock:

Hope you sort it.

philb
Posts: 118
Joined: Sat Aug 05, 2017 6:05 pm

Re: A3010: Bodge wires

Postby philb » Tue Sep 26, 2017 7:47 pm

BradBrad wrote:If its just something to do with the battery or floppy, can it run without the wire? As long as it is nothing to do with the cpu clock. I seem to recall reading that running it without a clock for more than a few seconds can screw the ARM chip? :shock:


The older ARMs were built from dynamic gates and required a continuously running clock for the logic to maintain its state. I suppose the concern is that if the clock stops and the gates lose their stored charge then you might end up with internal drive conflicts or low-impedance paths from power to ground. That's possibly more of a risk if the clock stops at some indeterminate logic level or if you apply multiple clock phases at once, rather than if the clock just freezes in what would otherwise be an instantaneously valid state. Simply freezing the clock would cause the state of the CPU to get scrambled after a while but I'm not sure that it would be likely to cause actual damage.

I think the Acorn documentation also warns that the DRAMs might be damaged if they are left powered without a clock, though this seems even more of a tenuous claim. The DRAMs themselves don't even receive the clock signal directly and I don't think I have ever heard of a DRAM suffering physical damage through not being refreshed.

The ARM6 and everything newer are fully static so this is a non-issue with those parts.


Return to “hardware”

Who is online

Users browsing this forum: No registered users and 5 guests